What is the Propagate-Then-Search pattern?

It is an algorithmic strategy that alternates between inferring necessary consequences of constraints and making educated guesses (search) to solve complex logic problems efficiently.

What is the Minimum Remaining Values (MRV) heuristic?

MRV is a strategy used during the search phase that selects the variable with the fewest possible legal values first, which typically leads to finding a solution or a contradiction much faster.

What are typical use cases for this Claude Code skill?

This skill is best used for Constraint Satisfaction Problems (CSPs) such as Sudoku solvers, classroom scheduling, map coloring, or any task where one choice restricts future options.

How does the 'fail fast' mechanism improve performance?

By propagating constraints immediately after every choice, the system detects contradictions as soon as they arise, allowing the algorithm to backtrack instantly instead of exploring dead-end branches.

Propagate-Then-Search Logic Pattern

Name: Propagate-Then-Search Logic Pattern
Author: jimmc414

byjimmc414

•

データサイエンスとML

Optimizes constraint satisfaction problem-solving by eliminating impossibilities through inference before initiating recursive search operations.

The Propagate-Then-Search skill implements a high-performance algorithmic pattern popularized by Peter Norvig for solving complex constraint-based challenges. It enables Claude to systematically approach puzzles, scheduling, and resource allocation tasks by first propagating constraints to prune the search space, drastically reducing the computational overhead of trial-and-error. By detecting contradictions early and utilizing heuristics like Minimum Remaining Values (MRV), this skill provides a robust framework for solving sophisticated logic problems where every assignment constrains future possibilities.

主な機能

01Immediate contradiction detection and fast-fail logic

02State isolation to prevent side-effect pollution during recursion

031 GitHub stars

04Minimum Remaining Values (MRV) heuristic for variable selection

05Constraint propagation for efficient search space pruning

06Inference-first approach to complex logic and resource puzzles

ユースケース

01Automating resource scheduling and personnel allocation

02Solving complex Sudoku, map coloring, or logic puzzles

03Optimizing constraint satisfaction problems in software architecture

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add jimmc414/claude-code-plugin-marketplace propagate-then-search

For use in Claude.ai and ChatGPT

Download Skill